Exploring the Difference in Diverse Schools' Destinies: A Research Note
by Erica Frankenberg - March 04, 2010
Many schools are diverse, but it is unclear whether they will remain so or are in the process of racial transition and appear diverse for a short period of time. This research note examines the stability (or instability) of racially diverse schools over the period of a decade. In particular, it studies the characteristics of schools and districts that may relate to whether schools remain stably diverse or whether they are susceptible to rapid racial change. Using federal education data, this analysis finds that school and district characteristics such as size and racial composition provide different contexts which may differentially affect the stability of diverse schools. These patterns have important implications for schools, teachers, and the communities in which they are located.
Public schools are in the midst of unprecedented racial transformation. In some regions of the country white students are no longer a majority of the public school enrollment. By 2006-07, in the largest metropolitan areas, more suburban students attended racially isolated nonwhite schools than segregated white schools, indicating patterns of minority suburbanization and segregation that have extended beyond the city lines. At the same time, legal and policy decisions are concurrently limiting student assignment plans seeking to integrate schools (e.g., Parents Involved, 2007), and when combined with the increasingly multiracial nature of the student enrollment, complex patterns of student composition are emerging. In addition to student racial composition, there may be other aspects of a schools demographic context to consider. In particular, neither research nor policy has considered the dimension of racial stability in schools. Are schools appearing diverse at one point in time and remaining so or do they only temporarily appear diverse in the midst of racial transition?
These patterns have important implications for schools, teachers, and the communities in which they are located since schools typically are the first sign of racial change (M. Orfield, 2002). An examination of educational and occupational outcomes of residents in stably diverse neighborhoods compared to diverse neighborhoods that are unstable found that they were higher in stably diverse Census tracts (G. Orfield, 1985). Racial stability portends substantial differences in the educational resources in a school such as the ability to retain high-quality, experienced teachers (Frankenberg, 2008; Jackson, 2009). Further, the stability of schools is important to consider due to the documented harms of racially isolated minority schools (Linn & Welner, 2007; Mickelson, 2009), which could result if diverse schools continue to lose white students.
This research note examines the stability (or instability) of racially diverse schools over the period of a decade. In particular, this note studies school and district demographic characteristics that may relate to whether schools remain stably diverse or whether they are susceptible to rapid racial change.
This note focuses on the dimension of school racial stability in diverse schools in addition to racial composition. Despite the differences in stably and transitioning diverse neighborhoods (Institute on Race and Poverty, 2006; G. Orfield, 1985), less is known about school transition, or even how to measure it (Morris, 2005; Real Estate Corp., 1976). Racial transition is defined here by change in a schools percentage of white students over a ten-year period. This definition stems from research suggesting that higher proportions of white students serve as a signal of school quality to families and teachers (Hanushek, Kain, & Rivkin, 2006; Holme, 2002; James, 1989; Mickelson, 2001; Pearce, 1980; Phillips & Chin, 2004; Reardon, Yun, & Eitle. 2000; Saporito, 2003). Earlier research has examined neighborhood or school change over differing periods of time, from as little as three years to twenty. A decade was chosen here to allow for racial change to become visible to community members and for possible corresponding actions by families to occur. While examining schools over a longer period of time might find that some rapidly transitioning schools stabilize, if school dynamics are similar to research on neighborhood or district patterns and family preferences, it suggests that schools may well only stabilize once they are no longer diverse (Charles, 2005; Institute on Race and Poverty, 2006; Ellen, 2000; Frankenberg, 2009b). Methodologically, as schools open and close for a number of reasons including population shifts, a disadvantage of studying schools over a longer period of time may result in a smaller percentage of schools being open at both points in time, which might bias the results.
This article uses demographic data from the National Center for Education Statistics Common Core of Data (CCD), which contains data from nearly every public school since the late 1980s. In order to evaluate racial change in addition to composition, I merged 1995-1996 and 2005-2006 CCD (n=74,193 matched schools for both years). In 1995, 1,361 schools did not match existing schools a decade later, indicating they had likely closed sometime prior to 2005-06. The average white percentage in these schools (58.0%) was lower than the average of matched schools (70.5%). In 2005-2006, 19,879 schools did not have matched data; their average white percentage was 52.7, lower than the average for matched schools (57.1%).
From 1995-1996 to 2005-2006, the percentage of white students in all public schools declined six points. I used the six-percentage point change as the norm to define categories of racial transition. In particular, rapid racially changing schools are those in which the white percentage declined more than three times the average rate during the decade (18%). Stably diverse schools are racially diverse (between 25% and 90% white in 2005-2006) with a decline in the percentage of white students between 0 and 12% during the decade. The definition of diverse was informed by theoretical considerations of critical mass and the percentage of white students in 2005-2006 (Clotfelter, 2004; Massey & Denton, 1988; Orfield, 1983; Pettigrew & Tropp, 2006). Although this definition of stability may mask fluctuation within the decade, a minority of schools increased in percentage of white students over the ten years. Even fewer (less than three percent of schools) increased in white percentage by six percentage points or more. The lack of large increases in percentage of white students makes it unlikely that a school would appear stable when measured only at the beginning and end of the decade but in reality it experienced significant mid-decade fluctuation.
Despite growing resegregation of public schools (e.g., Clotfelter, 2004; Orfield, 2009), a great many diverse schools exist although their prevalence varies substantially across the country. They are most common in the South and the West, the two most diverse regions of the country, where more than two out of three schools are classified as diverse. In no other region, however, were a majority of schools diverse. They were also most common in mid-sized cities, while only one-third of rural schools were diverse.
Nearly half of schools in 2005-2006 were classified as diverse, although only about half of these schools (n=18,393) were stably diverse when examining racial composition across the preceding decade. Another 10,663 schools experienced rapid racial transition over the decade examined. These numbers demonstrate that there are large numbers of diverse schools in both categories of racial stability.
RACIAL COMPOSITION OF SCHOOL
The differences in percentage of white students in the two types of diverse schools at the beginning of the decade were statistically different from one another: white students comprised 78% of students in a stably diverse school, on average, while rapidly transitioning schools contained a lower percentage of white students (64%) on average. Stably diverse schools had, on average, lower shares of black and Latino students than rapidly transitioning schools (each were approximately 6 to 7 percentage points lower on average).
Housing literature describes various tipping or turning points at which racial integration begins to recede because of the perception of too many nonwhite residents (Charles, 2005; Clark, 1987; Institute on Race and Poverty, 2006). For these purposes, the turning point adapted from the Institute on Race and Poverty (2006)s analysis is the point at which a school in 1995 is more likely to become a rapidly transitioning environment rather than remain stable.1 Although this analysis was also conducted for black and white percentage (available from author), schools were particularly sensitive to Hispanic students: the turning point for Hispanic percentage was 15% in 1995. Thus, schools with greater than 15% Hispanic students in 1995 were more likely to experience rapid transition rather than remain stably diverse. For reference, Figure 1 also includes the likelihood of becoming segregated white or segregated minority; schools with 45% or more Hispanic students were more likely to become segregated minority schools a decade later than be classified as either type of diverse school.
Figure 1. Distribution of Schools in 2005, by Percentage of Hispanic Students in School in 1995
Note: Lines reflect five percentage point moving averages for the Hispanic percentage in 1995 (e.g., along the x-axis).
Source: NCES Common Core of Data, Public School Universe, 1995-1996 and 2005-2006.
LOCATION OF SCHOOL
Research finds that the extent of segregation in racially isolated nonwhite schools, particularly for black and Latino students, differs according to whether students live in cities, suburbs, or rural areas (Frankenberg, 2008). Geographic differences, it turns out, are also prevalent among types of diverse schools.
Nearly two-fifths of stably diverse schools are in low-density areas (town or rural areas), where student mobility is less likely than in cities or some suburban areas. Fewer schools in these same locales are experiencing rapid racial transition (Table 1). In fact, many remain segregated white schools (Orfield, 2009). While stably diverse schools predominate in towns and rural areas, similar shares of rapidly changing schools are located in cities. The one place where there are morealmost twice as manyrapidly changing schools than stably diverse schools is in large central cities. There is also a much smaller percentage of stably integrated schools in all cities. Two possible explanations for this trend exist: either many urban districts have few white students remaining (Frankenberg, 2009a) or there is family mobility in and out of cities.
Table 1. School geographic location by racial stability of school, 2005-06
Note: Schools geographic location comes from the NCES Common Core of Datas locale code. Large central cities are defined as having at least 250,000 residents.
Source: Authors calculation of NCES Common Core of Data, Public School Universe, 1995-1996 and 2005-2006.
Relatively equal percentages of stably diverse and rapidly transitioning schools are in suburban areas, although they differ by the size of metropolitan areas. Suburbs in the largest metropolitan areas had the highest concentration of rapidly changing schools. All told, more than half of rapidly changing schools are in the largest metropolitan areas, areas which may be centers of immigration and residential transition that expand from urban cores into first-ring suburbs and beyond (Frey, 2001). Stably diverse schools, however, are relatively evenly split between metropolitan areas, regardless of size, although in larger metros, most of these schools are in the suburbs.
REGION OF THE COUNTRY
Racial isolation of black and Latino students varies substantially between states and regions (Boger & Orfield, 2005; Clotfelter, 2004; G. Orfield, 2009). Black and Latino students are especially isolated in schools in the Northeast and Midwest, which may be partially due to the municipal structure of school district boundaries in these regions metropolitan areas (G. Orfield, 2001).
While the Northeast and Midwest have the lowest percentage of schools that are diverse, the two regions differ in terms of the stability of their diverse schools. Overall, the Midwest has more diverse schools. Yet they have twice as many rapidly transitioning schools (2,080) than the Northeast (1,012) and only have 25% more stably diverse schools (3,625 in the Midwest, 2,889 in the Northeast). In other words, diverse schools in the Midwest are more likely to be in the midst of racial transition. Diverse schools in the Northeast are more likely than Midwestern schools to be stably diverse, perhaps because of the relatively slow racial change in the region over the last 15 years. Since 1990, the percentage of white public school students in the Northeast has declined seven [percentage] points (G. Orfield & Lee, 2007), which might suggest that while there has been a decline of white students in the regions it has occurred slowly enough that schools have been able to maintain their diversity during the last decade.
The South and the West are the most diverse regions of the country, and they also have the highest percentages of diverse schools. In particular, they have higher percentages of rapidly transitioning schools (18.0% and 20.1% of schools in the region, respectively) than do other regions of the country, which may be due, in part, to the dissolving of desegregation plans in the South (Clotfelter, Ladd, & Vidgor, 2005) and demographic transition in the West (G. Orfield & Lee, 2007).
DISTRICT RACIAL CONTEXT
School district boundaries are another important influence on school racial context. Clotfelter (2004) estimated that 70% of existing segregation is due to segregation between districts, with the remaining portion a result of within-district segregation (see also Reardon & Yun, 2005). This is due to deeply entrenched notions of local control along with legal rulings that largely restrict inter-district mandatory desegregation plans; boundary lines themselves may further sort populations (Frankenberg, 2009b; Weiher, 1991). In some districts, there may be few white--or few nonwhite--students and therefore few diverse schools (Frankenberg, 2009a).
Mirroring school racial composition differences, diverse schools that remained stable are located in districts that at the beginning of the decade had a higher share of white students--nearly three out of four are white--and a lower percentage of both black and Latino students. Further, while a majority of both types of schools were in districts that were predominantly white in 1995, almost all stably diverse schools were in majority white districts in 1995 (89%). A lesser share of rapidly transitioning schools (71%) was located in majority white districts. This suggests that stably diverse schools, by virtue of their districts racial composition in 1995, may have been more poised to retain a diverse student body.
Table 2. District Racial Composition of Diverse Schools in 1995
Source: From authors calculation of NCES Common Core of Data, Public School Universe, 1995-1996 and 2005-2006.
SIZE OF DISTRICTS
Larger districts may, theoretically, be able to better weather a communitys racial transition if they, for example, encompass both a city and a suburb in the same district (G. Orfield, 2001; see also Weiher, 1991). Most residential moves in such cases would still keep families within the districts boundaries. While this might help districts remain stable, it is unknown how this might affect school stability within districts.
Among this group of diverse schools, we see a different pattern that may suggest that district size affects schools in a more nuanced way. In this analysis, it is unknown whether larger districts are large municipal city districts, large suburban ones, or perhaps contain city and suburban parts of a metropolitan area. Larger districts had a higher proportion of diverse schools, although there were significant differences by district size in whether these diverse schools were stable. The mean district size in 1995 for rapidly transitioning schools (34,611) was significantly higher than that of stably diverse schools (17,593)almost twice as large, on average. Among school districts with at least 10,000 students, more diverse schools experienced rapid transition than those that were stable. Although both types of diverse schools are under-represented among the nations smallest districts, larger district size may be associated with other characteristics (e.g., immigration) that make diverse schools more susceptible to racial transition.2 This also suggests that the specific type of district (e.g., countywide or single municipality) may play an important role in racial stability.
This note has explored the added dimension of racial stability along with schools racial composition. Despite growing segregation, diverse schools in 2005-06 enrolled more than 28 million students. While thousands of schools have remained stably diverse over the last decade, thousands of other diverse schools are experiencing rapid transition, which threatens their ability to remain diverse. Though variety exists within each category of diverse schools, these findings suggest strong variation between categories of stably and unstably diverse schools on school and district characteristics.
Schools student composition may act as signaling devices to families making schooling decisions, and given the different tolerances by race for diverse environments (Charles, 2005), diverse schools may be particularly sensitive to perceptions of shifting demographics. Social science evidence suggests that if schools are identifiable or unbalanced they are likely to become more so over time (Gaertner & Dovidio, 1986; Henig, 1996; Brief of American Psychological Association, 2006). As parents seek to avoideven if unconsciouslytransition, any modest racial transition may accelerate as schools or districts are unable to recruit students and families similar to those leaving (Weiher, 1991). There may also be anxiety among the school community at the rapidity of the changes that are occurring, further destabilizing the community by a loss of teachers (Morris, 2005).
One of the most intriguing findings is the sensitivity among diverse schools to Hispanic percentage at the beginning of the decade examined. At just 15% of Hispanic students in 1995, more schools a decade later were rapidly transitioning than stably diverse. This finding is similar to those in the residential context finding low tolerance for black and Hispanic students in diverse Census tracts (IRP, 2006). Schools remaining stably diverse had a low tolerance for Hispanic students in 1995.
Likewise, regional and geographic variation relates to whether schools remain stably diverse or whether they are more likely to experience rapid transition. In particular, cities and more diverse regions, the West and South, had higher shares of diverse schools that were rapidly transitioning. As these areas already have high levels of minority segregation, this trend may indicate a further rise in such patterns in the years to come if these schools are not stabilized.
Similar to other research on school segregation, this analysis finds that district context relates to the racial stability of diverse schools. The districts of stably diverse schools had significantly higher percentages of white students at the beginning of the decade examined than those of rapidly racially changing schools. Stably diverse schools were also in significantly smaller districts than rapidly changing schools, which amplifies previous literature about the relationship of district size and racial stability. The differences in district contexts in 2005 for diverse schoolsrapidly transitioning schools are even more likely in 2005 to be in majority nonwhite, larger districtssuggest that these types of diverse schools may diverge further.
Of course, each of these school and district characteristics may also reflect variations in other characteristics, such as the housing market, that relate to patterns discussed here. Future research should investigate the relationship of non-educational characteristics with stably and rapidly transitioning diverse schools.
The implications of these trends are likely to be significant. If racial transition continues unabated in rapidly changing diverse schools, the number of racially isolated minority schools will grow. With research finding that such schools tend to have lower graduation rates, these demographic trends portend limited opportunity for students who attend them and have consequences for our society as well. Further, students in racially isolated minority schools will miss the opportunity to prepare for citizenship in an increasingly diverse society. As a result, districts and communities should monitor the added dimension of racial stability and take efforts to stabilize diverse schools to maintain benefits for students who attend such schools.
1. The turning point here was calculated by plotting the percentage of schools in each diversity category in 2005 at each value of Hispanic percentage in the school in 1995. It was smoothed using a five-percentage point moving average for the x-axis. Similar turning points, not shown here, were calculated for black and white percentage in school in 1995.
2. Unlike districts, school size in 1995 is not significantly related to whether diverse schools are racially stable or transitioning.
Boger, J. C., & Orfield, G. (2005). School resegregation: Must the south turn back? Chapel Hill, NC: University of North Carolina Press.
Brief of the American Psychological Association as Amicus Curiae in Support of Respondents, Parents Involved in Community Schools v. Seattle School District No. 1, et al. and Crystal D. Meredith v. Jefferson County Board of Education et al., (Nos. 05-908 & 05-915) (Sp. Ct. 2006).
Charles, C. Z. (2005). Can we live together? Racial preferences and neighborhood outcomes. In X. d. S. Briggs (Ed.). The geography of opportunity: Race and housing choice in metropolitan America (pp. 45-80). Washington, DC: Brookings Institution Press.
Clark, T.A. (1987). The suburbanization process and residential segregation. In G. A. Tobin (Ed.), Dividing neighborhoods: Changing patterns of racial segregation (pp. 115-137). Urban affairs annual reviews, vol. 32. Newbury Park, CA: Sage Publications.
Clotfelter, C. T. (2004). After Brown: The rise and retreat of school desegregation. Princeton: Princeton University Press.
Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2005). Federal oversight, local control, and the specter of resegregation in southern schools. Cambridge, MA: National Bureau of Economic Research Working Paper.
Ellen, I. G. (2000). Sharing Americas neighborhoods: The prospects for stable racial integration. Boston, MA: Harvard University Press.
Frankenberg, E. (2009a). The demographic context of urban schools and districts. Equity and Excellence in Education 42(3): 255-271.
Frankenberg, E. (2009b). Splintering school districts: Understanding the link between segregation and fragmentation. Law and Social Inquiry 34 (4): 869-909.
Frankenberg, E. (2008). America's diverse, racially changing schools and their teachers. Ed.D. dissertation, Harvard University.
Frey, W. H. (2001). Melting pot suburbs: A census 2000 study of suburban diversity. Census 2000 Series. Washington, D.C.: The Brookings Institution Center on Urban and Metropolitan Policy.
Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (2009). New evidence about Brown v. Board of Education: The complex effects of school racial composition on achievement. Journal of Labor Economics, 27(3), 349-384.
Holme, J. J. (2002). Buying homes, buying schools: School choice and the social construction of school quality. Harvard Educational Review, 177-205.
Institute on Race and Poverty (IRP). (2006). Minority suburbanization, stable integration, and economic opportunity in fifteen metropolitan regions. Minneapolis: University of Minnesota Law School.
Jackson, C. K. (2009). Student demographics, teacher sorting, and teacher quality: Evidence from the end of school desegregation. Journal of Labor Economics, 27(2), 213-256.
James, D. R. (1989). City limits on racial equality: The effects of city-suburb boundaries on public-school desegregation, 1968-1976. American Sociological Review, 54(6), 963-985.
Linn, R. L., & Welner, K. G. (Eds.). (2007). Race-conscious policies for assigning students to schools: Social science research and the Supreme Court cases. Washington, DC: National Academy of Education.
Massey, D. S. & Denton, N.A. (1988). The dimensions of racial segregation. Social Forces, 67, 281-315.
Mickelson, R. A. (2001). Subverting Swann: First- and second- generation segregation in Charlotte, North Carolina. American Educational Research Journal, 38(2), 215-252.
Mickelson, R.A. (2009). Twenty-first century social science on school racial diversity and educational outcomes. Ohio State Law Journal, 69(6),1173-1228.
Morris, E. W. (2005). From middle class to trailer trash: Teachers perceptions of white students in a predominately minority school. Sociology of Education, 78(2), 99-121.
Orfield, G. (1983). Public school desegregation in the United States, 1968-1980. Washington, D.C.: Joint Center for Political Studies.
Orfield, G. (1985). Ghettoization and its alternatives. In P. Peterson (Ed.), The New Urban Reality. Washington, D.C.: Brookings Institution Press.
Orfield, G. (2001). Metropolitan school desegregation: Impacts on metropolitan society. In j. a. powell, G. Kearney, & V. Kay (Eds.), In pursuit of a dream deferred (pp. 121-157). Peter Lang: New York.
Orfield, G. (2009). Reviving the goal of an integrated society: A 21st century challenge. Los Angeles: Civil Rights Project.
Orfield, G., & Lee, C. (2007). Historic reversals, accelerating resegregation, and the need for new integration strategies. Los Angeles: The Civil Rights Project.
Orfield, M. (2002). American metropolitics: The new suburban reality. Washington, D.C.: Brookings Institution.
Parents Involved in Community Schools v. Seattle School District No.1 and Meredith v. Jefferson County Board of Education, 127 U.S. 2738 (2007).
Pearce, D. (1980). Breaking down the barriers: New evidence on the impact of metropolitan school desegregation on housing patterns. Washington D.C.: National Institute of Education.
Pettigrew, T., & Tropp, L. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90, 751-783.
Phillips, M., & Chin, T. (2004). School inequality: What do we know? In K. Neckerman (Ed.), Social inequality (pp. 467-519). New York: Russell Sage Foundation.
Real Estate Research Corp. (1976). Stability of racial mix in Illinois schools. Springfield: Illinois State Office of Education.
Reardon, S. F., & Yun, J. T. (2005). Integrating neighborhoods, segregating schools: The retreat from school desegregation in the South, 1990-2000. In J. C. Boger & G. Orfield (Eds.), School resegregation: Must the South turn back? (pp. 51-69). Chapel Hill: University of North Carolina Press.
Reardon, S.F., Yun, J.T., & Eitle, T.M. (2000). The changing structure of school segregation: Measurement and evidence of multi-racial metropolitan school segregation, 1989-1995. Demography, 37(3), 351-364.
Saporito, S. (2003). Private choices, public consequences: Magnet school choice and segregation by race and poverty. Social Problems, 50(2), 181-203.
Weiher, G. R. (1991). The fractured metropolis: Political fragmentation and metropolitan segregation. Albany: State University of New York Press.